Accuracy Of Landmark Detection Research Articles

Fully automated planning for anatomical fetal brain MRI on 0.55T.

Widening the availability of fetal MRI with fully automatic real-time planning of radiological brain planes on 0.55T MRI. Deep learning-based detection of key brain landmarks on a whole-uterus echo planar imaging scan enables the subsequent fully automatic planning of the radiological single-shot Turbo Spin Echo acquisitions. The landmark detection pipeline was trained on over 120 datasets from varying field strength, echo times, and resolutions and quantitatively evaluated. The entire automatic planning solution was tested prospectively in nine fetal subjects between 20 and 37 weeks. A comprehensive evaluation of all steps, the distance between manual and automatic landmarks, the planning quality, and the resulting image quality was conducted. Prospective automatic planning was performed in real-time without latency in all subjects. The landmark detection accuracy was 4.2 2.6 mm for the fetal eyes and 6.5 3.2 for the cerebellum, planning quality was 2.4/3 (compared to 2.6/3 for manual planning) and diagnostic image quality was 2.2 compared to 2.1 for manual planning. Real-time automatic planning of all three key fetal brain planes was successfully achieved and will pave the way toward simplifying the acquisition of fetal MRI thereby widening the availability of this modality in nonspecialist centers.

PELE scores: pelvic X-ray landmark detection with pelvis extraction and enhancement.

Pelvic X-ray (PXR) is widely utilized in clinical decision-making associated with the pelvis, the lower part of the trunk that supports and balances the trunk. In particular, PXR-based landmark detection facilitates downstream analysis and computer-assisted diagnosis and treatment of pelvic diseases. Although PXR has the advantages of low radiation and reduced cost compared to computed tomography (CT), it characterizes the 2D pelvis-tissue superposition of 3D structures, which may affect the accuracy of landmark detection in some cases. However, the superposition nature of PXR is implicitly handled by existing deep learning-based landmark detection methods, which mainly design the deep network structures for better detection performances. Explicit handling of the superposition nature of PXR is rarely done. In this paper, we explicitly focus on the superposition of X-ray images. Specifically, we propose a pelvis extraction (PELE) module that consists of a decomposition network, a domain adaptation network, and an enhancement module, which utilizes 3D prior anatomical knowledge in CT to guide and well isolate the pelvis from PXR, thereby eliminating the influence of soft tissue for landmark detection. The extracted pelvis image, after enhancement, is then used for landmark detection. We conduct an extensive evaluation based on two public and one private dataset, totaling 850 PXRs. The experimental results show that the proposed PELE module significantly improves the accuracy of PXRs landmark detection and achieves state-of-the-art performances in several benchmark metrics. The design of PELE module can improve the accuracy of different pelvic landmark detection baselines, which we believe is obviously conducive to the positioning and inspection of clinical landmarks and critical structures, thus better serving downstream tasks. Our project has been open-sourced at https://github.com/ECNUACRush/PELEscores .

Automated anatomical landmark detection on 3D facial images using U-NET-based deep learning algorithm.

Facial anthropometry based on 3-dimensional (3D) imaging technology, or 3D photogrammetry, has gained increasing popularity among surgeons. It outperforms direct measurement and 2-dimensional (2D) photogrammetry because of many advantages. However, a main limitation of 3D photogrammetry is the time-consuming process of manual landmark localization. To address this problem, this study developed a U-NET-based deep learning algorithm to enable automated and accurate anatomical landmark detection on 3D facial models. The main structure of the algorithm stacked 2 U-NETs. In each U-NET block, we used 3×3 convolution kernel and rectified linear unit (ReLU) as activation function. A total of 200 3D images of healthy cases, acromegaly patients, and localized scleroderma patients were captured by Vectra H1 handheld 3D camera and input for algorithm training. The algorithm was tested to detect 20 landmarks on 3D images. Percentage of correct key points (PCK) and normalized mean error (NME) were used to evaluate facial landmark detection accuracy. Among healthy cases, the average NME was 1.4 mm. The PCK reached 90% when the threshold was set to the clinically acceptable limit of 2 mm. The average NME was 2.8 and 2.2 mm among acromegaly patients and localized scleroderma patients, respectively. This study developed a deep learning algorithm for automated facial landmark detection on 3D images. The algorithm was innovatively validated in 3 different groups of participants. It achieved accurate landmark detection and improved the efficiency of 3D image analysis.

Craniomaxillofacial Bone Segmentation and Landmark Detection Using Semantic Segmentation Networks and an Unbiased Heatmap.

Craniomaxillofacial (CMF) surgery always relies on accurate preoperative planning to assist surgeons, and automatically generating bone structures and digitizing landmarks for CMF preoperative planning is crucial. Since the soft and hard tissues of the CMF regions possess complicated attachment, segmenting the CMF bones and detecting the CMF landmarks are challenging problems. In this study, we proposed a semantic segmentation network to segment the maxilla, mandible, zygoma, zygomatic arch, and frontal bones. Then, we obtained the minimum bounding box around the CMF bones. After cropping, we used the top-down heatmap landmark detection network, similar to the segmentation module, to identify 18 CMF landmarks from the cropping patch. In addition, an unbiased heatmap encoding method was proposed to generate actual landmark coordinates in the heatmap. To overcome quantization effects in the heatmap-based landmark detection networks, the distribution-prior coordinate representation of medical landmarks (DCRML) was proposed to utilize the prior distribution of the encoding heatmap, approximating the accurate landmark coordinates in heatmap decoding by Taylor's theorem. The encoding and decoding method can easily contribute to other existing landmark detection frameworks based on heatmaps; consequently, these approaches can readily benefit without changing model structure. We used prior segmentation knowledge to enhance the semantic information around the landmarks, increasing landmark detection accuracy. The proposed framework was evaluated by 100 healthy persons and 86 patients from multicenter cooperation. The mean Dice score of our proposed segmentation network achieved over 88 %; in particular, the mandible accuracy was approximately 95%. The mean error of landmarks was 1.84 ±1.32 mm.

Lateral Cephalometric Landmark Annotation Using Histogram Oriented Gradients Extracted from Region of Interest Patches.

Two-dimensional cephalometric image analysis plays a crucial role in orthodontic diagnosis and treatment planning. While deep learning-based algorithms have emerged to automate the laborious task of anatomical landmark annotation, their effectiveness is hampered by the challenges of acquiring and labelling clinical data. In this study, we propose a model that leverages conventional machine learning techniques to enhance the accuracy of landmark detection using limited dataset. Our methodology involves coarse localization through region of interest (ROI) extraction and fine localization utilizing histogram-oriented gradient (HOG) feature. The image patch containing landmark pixels is classified using the light gradient boosting machine (LGBM) algorithm. To evaluate our model's performance, we conducted rigorous tests on the ISBI Cephalometric dataset and Dental Cepha dataset, aiming to achieve accuracy within a 2mm radial precision range. We also employed cross-validation to assess our approach, providing a robust evaluation. Our model's performance on the ISBI Cephalometric dataset showed an accuracy rate of 77.11% within the desired 2 mm radial precision range. The cross-validation results further confirmed the effectiveness of our approach, yielding a mean accuracy of 78.17%. Additionally, we applied our model to the Dental Cepha dataset, where we achieved a remarkable landmark detection accuracy of 84%. The results demonstrate that traditional machine learning techniques can be effective for accurate landmark detection in cephalometric images, even with limited data. Our findings highlight the potential of these techniques for clinical applications, where large datasets of labelled images may not be available.

Integrating biplane information and context for spine landmark detection.

Spine landmark detection is of great significance for spinal morphological parameter assessment and three-dimensional reconstruction of the human spine. This detection task generally involves locating spine landmarks in the anterior-posterior (AP) and lateral (LAT) X-rays of the spine. Recently, the two-stage methods for AP spine landmark detection achieve better performance. However, these methods perform poorly in LAT landmark detection because of poor detection accuracy of LAT vertebra due to occlusion. To solve this problem, this paper proposes a new two-stage spine landmark detection method. In the first stage, this paper propose a biplane vertebra detection network for vertebra detection on AP and X-rays simultaneously. Then an epipolar module and a context enhancement module are proposed to assist LAT vertebra detection by using the biplane information and the context information of the vertebrae respectively. In the second stage, the landmarks can be obtained in the detected vertebrae area. Extensive experiment results conducted on a dataset containing 328 pairs of X-rays demonstrate that our method improves the vertebra and landmark detection accuracy.

Facial Landmark, Head Pose, and Occlusion Analysis Using Multitask Stacked Hourglass

In this study, we proposed a multitask network architecture for three attributes, landmark, head pose, and occlusion, from a face image. A 2-stacked hourglass with three task-specific heads is the proposed network architecture. We also designed three auxiliary components for the network. First is the feature pyramid fusion module, which plays a crucial role in facilitating contextual information from various receptive fields. Second is the interlevel occlusion-aware fusion module, which explicitly fuses intermediate occlusion prediction between subnetworks. The third is the gimbal-lock-free head pose head, which outputs a rotation matrix from a 6D rotation representation. We conducted an ablative study of these auxiliary components to determine their impacts on the network. Additionally, we introduced the landmark heatmap scaling approach to avoid falling local minima. We trained the proposed network with a 300W-LP dataset for landmark and head pose and a C-CM dataset for occlusion. Then, we fine-tuned the network using the 300W or WFLW dataset, instead of the 300W-LP dataset for the landmark task. This 2-stage training method contributes to enhancing the landmark detection accuracy and that of other tasks. In the experiments, we assessed the proposed network using eight test datasets and task-specific metrics.

Anthropometric Landmarking for Diagnosis of Cranial Deformities: Validation of an Automatic Approach and Comparison with Intra- and Interobserver Variability.

Shape analysis of infant's heads is crucial to diagnose cranial deformities and evaluate head growth. Currently available 3D imaging systems can be used to create 3D head models, promoting the clinical practice for head evaluation. However, manual analysis of 3D shapes is difficult and operator-dependent, causing inaccuracies in the analysis. This study aims to validate an automatic landmark detection method for head shape analysis. The detection results were compared with manual analysis in three levels: (1) distance error of landmarks; (2) accuracy of standard cranial measurements, namely cephalic ratio (CR), cranial vault asymmetry index (CVAI), and overall symmetry ratio (OSR); and (3) accuracy of the final diagnosis of cranial deformities. For each level, the intra- and interobserver variability was also studied by comparing manual landmark settings. High landmark detection accuracy was achieved by the method in 166 head models. A very strong agreement with manual analysis for the cranial measurements was also obtained, with intraclass correlation coefficients of 0.997, 0.961, and 0.771 for the CR, CVAI, and OSR. 91% agreement with manual analysis was achieved in the diagnosis of cranial deformities. Considering its high accuracy and reliability in different evaluation levels, the method showed to be feasible for use in clinical practice for head shape analysis.

Semi-supervised anatomical landmark detection via shape-regulated self-training

Well-annotated medical images are costly and sometimes even impossible to acquire, hindering landmark detection accuracy to some extent. Semi-supervised learning alleviates the reliance on large-scale annotated data by exploiting the unlabeled data to understand the population structure of anatomical landmarks. The global shape constraint is the inherent property of anatomical landmarks that provides valuable guidance for more consistent pseudo labelling of the unlabeled data, which is ignored in the previously semi-supervised methods. In this paper, we propose a model-agnostic shape-regulated self-training framework for semi-supervised landmark detection by fully considering the global shape constraint. Specifically, to ensure pseudo labels are reliable and consistent, a PCA-based shape model adjusts pseudo labels and eliminate abnormal ones. A novel Region Attention loss to make the network automatically focus on the structure consistent regions around pseudo labels. Extensive experiments show that our approach outperforms other semi-supervised methods and achieves the relative improvement of 3.8%, 6.1% and 6.3% on three medical image datasets. Furthermore, our framework is flexible and can be used as a plug-and-play module integrated into most superviseüd methods to improve performance further.

Fully automated Assessment of Knee Alignment from Full-Leg X-Rays employing a “YOLOv4 And Resnet Landmark regression Algorithm” (YARLA): Data from the Osteoarthritis Initiative

Background and ObjectiveWe present a fully automated method for the quantification of knee alignment from full-leg radiographs. MethodsA state-of-the-art object detector, YOLOv4, was trained to locate regions of interests in full-leg radiographs for the hip joint, knee, and ankle. Residual neural networks were trained to regress landmark coordinates for each region of interest. Based on the detected landmarks the knee alignment, i.e., the hip-knee-ankle (HKA) angle was computed. The accuracy of landmark detection was evaluated by a comparison to manually placed ones for 180 radiographs. The accuracy of HKA angle computations was assessed on the basis of 2,943 radiographs by a comparison to results of two independent image reading studies (Cooke; Duryea) both publicly accessible via the Osteoarthritis Initiative. The agreement was evaluated using Spearman’s Rho, weighted kappa, and regarding the correspondence of the class assignment. ResultsThe average deviation of landmarks manually placed by experts and automatically detected ones by our proposed “YOLOv4 And Resnet Landmark regression Algorithm” (YARLA) was less than 2.0 ± 1.5 mm for all structures. The average mismatch between HKA angle determinations of Cooke and Duryea was 0.09 ± 0.63°; YARLA resulted in a mismatch of 0.09 ± 0.73° compared to Cooke and of 0.18 ± 0.67° compared to Duryea. Cooke and Duryea agreed almost perfectly with respect to a weighted kappa value of 0.86, and showed an excellent reliability as measured by a Spearman’s Rho value of 0.98. Similar values were achieved by YARLA, i.e., a weighted kappa value of 0.83 and 0.87 and a Spearman’s Rho value of 0.98 and 0.98 compared to Cooke and Duryea, respectively. Cooke and Duryea agreed in 91% of all class assignments and YARLA did so in 90% against Cooke and 92% against Duryea. ConclusionsYARLA yields HKA angles similar to those of human experts and provides a basis for an automated assessment of knee alignment in full-leg radiographs.

CircleNet for Hip Landmark Detection

Landmark detection plays a critical role in diagnosis of Developmental Dysplasia of the Hip (DDH). Heatmap and anchor-based object detection techniques could obtain reasonable results. However, they have limitations in both robustness and precision given the complexities and inhomogeneity of hip X-ray images. In this paper, we propose a much simpler and more efficient framework called CircleNet to improve the accuracy of landmark detection by predicting landmark and corresponding radius. Using the CircleNet, we not only constrain the relationship between landmarks but also integrate landmark detection and object detection into an end-to-end framework. In order to capture the effective information of the long-range dependency of landmarks in the DDH image, here we propose a new context modeling framework, named the Local Non-Local (LNL) block. The LNL block has the benefits of both non-local block and lightweight computation. We construct a professional DDH dataset for the first time and evaluate our CircleNet on it. The dataset has the largest number of DDH X-ray images in the world to our knowledge. Our results show that the CircleNet can achieve the state-of-the-art results for landmark detection on the dataset with a large margin of 1.8 average pixels compared to current methods. The dataset and source code will be publicly available.

Deep, Landmark-Free FAME: Face Alignment, Modeling, and Expression Estimation

We present a novel method for modeling 3D face shape, viewpoint, and expression from a single, unconstrained photo. Our method uses three deep convolutional neural networks to estimate each of these components separately. Importantly, unlike others, our method does not use facial landmark detection at test time; instead, it estimates these properties directly from image intensities. In fact, rather than using detectors, we show how accurate landmarks can be obtained as a by-product of our modeling process. We rigorously test our proposed method. To this end, we raise a number of concerns with existing practices used in evaluating face landmark detection methods. In response to these concerns, we propose novel paradigms for testing the effectiveness of rigid and non-rigid face alignment methods without relying on landmark detection benchmarks. We evaluate rigid face alignment by measuring its effects on face recognition accuracy on the challenging IJB-A and IJB-B benchmarks. Non-rigid, expression estimation is tested on the CK+ and EmotiW’17 benchmarks for emotion classification. We do, however, report the accuracy of our approach as a landmark detector for 3D landmarks on AFLW2000-3D and 2D landmarks on 300W and AFLW-PIFA. A surprising conclusion of these results is that better landmark detection accuracy does not necessarily translate to better face processing. Parts of this paper were previously published by Tran et al. (2017) and Chang et al. (2017, 2018).

AMID: Accurate Magnetic Indoor Localization Using Deep Learning.

Geomagnetic-based indoor positioning has drawn a great attention from academia and industry due to its advantage of being operable without infrastructure support and its reliable signal characteristics. However, it must overcome the problems of ambiguity that originate with the nature of geomagnetic data. Most studies manage this problem by incorporating particle filters along with inertial sensors. However, they cannot yield reliable positioning results because the inertial sensors in smartphones cannot precisely predict the movement of users. There have been attempts to recognize the magnetic sequence pattern, but these attempts are proven only in a one-dimensional space, because magnetic intensity fluctuates severely with even a slight change of locations. This paper proposes accurate magnetic indoor localization using deep learning (AMID), an indoor positioning system that recognizes magnetic sequence patterns using a deep neural network. Features are extracted from magnetic sequences, and then the deep neural network is used for classifying the sequences by patterns that are generated by nearby magnetic landmarks. Locations are estimated by detecting the landmarks. AMID manifested the proposed features and deep learning as an outstanding classifier, revealing the potential of accurate magnetic positioning with smartphone sensors alone. The landmark detection accuracy was over 80% in a two-dimensional environment.

Learning Pose-Aware Models for Pose-Invariant Face Recognition in the Wild.

We propose a method designed to push the frontiers of unconstrained face recognition in the wild with an emphasis on extreme out-of-plane pose variations. Existing methods either expect a single model to learn pose invariance by training on massive amounts of data or else normalize images by aligning faces to a single frontal pose. Contrary to these, our method is designed to explicitly tackle pose variations. Our proposed Pose-Aware Models (PAM) process a face image using several pose-specific, deep convolutional neural networks (CNN). 3D rendering is used to synthesize multiple face poses from input images to both train these models and to provide additional robustness to pose variations at test time. Our paper presents an extensive analysis of the IARPA Janus Benchmark A (IJB-A), evaluating the effects that landmark detection accuracy, CNN layer selection, and pose model selection all have on the performance of the recognition pipeline. It further provides comparative evaluations on IJB-A and the PIPA dataset. These tests show that our approach outperforms existing methods, even surprisingly matching the accuracy of methods that were specifically fine-tuned to the target dataset. Parts of this work previously appeared in[1] and[2].

Fully automated quantitative cephalometry using convolutional neural networks.

Quantitative cephalometry plays an essential role in clinical diagnosis, treatment, and surgery. Development of fully automated techniques for these procedures is important to enable consistently accurate computerized analyses. We study the application of deep convolutional neural networks (CNNs) for fully automated quantitative cephalometry for the first time. The proposed framework utilizes CNNs for detection of landmarks that describe the anatomy of the depicted patient and yield quantitative estimation of pathologies in the jaws and skull base regions. We use a publicly available cephalometric x-ray image dataset to train CNNs for recognition of landmark appearance patterns. CNNs are trained to output probabilistic estimations of different landmark locations, which are combined using a shape-based model. We evaluate the overall framework on the test set and compare with other proposed techniques. We use the estimated landmark locations to assess anatomically relevant measurements and classify them into different anatomical types. Overall, our results demonstrate high anatomical landmark detection accuracy ([Formula: see text] to 2% higher success detection rate for a 2-mm range compared with the top benchmarks in the literature) and high anatomical type classification accuracy ([Formula: see text] average classification accuracy for test set). We demonstrate that CNNs, which merely input raw image patches, are promising for accurate quantitative cephalometry.

Improving landmark detection accuracy for self-localization through baseboard recognition

Acquiring information related to the surrounding environment and environmental mapping are important issues for realizing autonomous robot movement in an unknown environment, and these processes require position estimations for self-localization. We have conducted previous research related to self-localization in indoor environments through baseboard recognition. In that method, a mobile robot performs self-localization using camera-acquired images, from which the intersection of baseboards and the vertical lines of attached doorframes are registered as landmarks. However, this method is limited in that self-localization cannot be performed in cases where baseboards are not partially available or where walls and baseboards are the same color. The present research aims at addressing these issues and implementing the method in an actual mobile robot. The implementing robot is allowed to move autonomously, performing mapping and self-localization in real time. Addressing the problem of same-colored walls requires identification of the border between the wall and the baseboard, so we propose an image processing method for identifying these lines. The problem of lack of baseboards is solved through supplemental use of odometry, which does not rely on visual characteristics. The proposed method is experimentally verified through use of our algorithm to recognize baseboards, perform self-localization, and create maps. The results verify that baseboards can be recognized even when they are the same color as walls. We furthermore verify that landmark-mapping performance when moving in hallways is improved over the previous version.

The Accuracy of Locating Lumbar Vertebrae When Using Palpation Versus Ultrasonography

ObjectiveThe purpose of this study was to determine the accuracy of locating lumbar vertebrae using palpation vs ultrasonography. MethodsIn this study, ultrasonic imaging was used by 2 experienced clinicians to identify the third lumbar spinous process (target) of a female participant. The target was then located by 16 undergraduate chiropractic students using clinical palpation techniques learned in their academic program (with participant seated and prone) and ultrasonic imaging learned through a 5-minute training video. Presumed target locations identified by students were recorded by infrared motion capture equipment. The coordinates of the presumed target site were then compared statistically. ResultsThere was no significant difference between the presumed target position identified by the students using sitting and prone palpation (P = .346). These positions were significantly different from the target location identified by expert clinicians using ultrasonic imaging (P < .0001 in both cases). The vertebra identified by ultrasonic imaging by the students was the same vertebra identified by the expert clinicians using ultrasound. This position error in the vertebra identified by palpation resulted in the students mistakenly identifying the L4 spinous process as the target vertebra. ConclusionsThis study found that ultrasonography provided more accurate identification of a lumbar spinal landmark when compared with palpation. In addition, our data suggest that ultrasonic imaging to identify spinal landmarks can be learned easily and can improve accuracy of landmark detection. Although the time to use ultrasonic imaging was greater than with palpation, these results suggest that this procedure could potentially be used in clinical practice to identify spinal landmarks.

Collaborative regression-based anatomical landmark detection

Anatomical landmark detection plays an important role in medical image analysis, e.g. for registration, segmentation and quantitative analysis. Among the various existing methods for landmark detection, regression-based methods have recently attracted much attention due to their robustness and efficiency. In these methods, landmarks are localised through voting from all image voxels, which is completely different from the classification-based methods that use voxel-wise classification to detect landmarks. Despite their robustness, the accuracy of regression-based landmark detection methods is often limited due to (1) the inclusion of uninformative image voxels in the voting procedure, and (2) the lack of effective ways to incorporate inter-landmark spatial dependency into the detection step. In this paper, we propose a collaborative landmark detection framework to address these limitations. The concept of collaboration is reflected in two aspects. (1) Multi-resolution collaboration. A multi-resolution strategy is proposed to hierarchically localise landmarks by gradually excluding uninformative votes from faraway voxels. Moreover, for informative voxels near the landmark, a spherical sampling strategy is also designed at the training stage to improve their prediction accuracy. (2) Inter-landmark collaboration. A confidence-based landmark detection strategy is proposed to improve the detection accuracy of ‘difficult-to-detect’ landmarks by using spatial guidance from ‘easy-to-detect’ landmarks. To evaluate our method, we conducted experiments extensively on three datasets for detecting prostate landmarks and head & neck landmarks in computed tomography images, and also dental landmarks in cone beam computed tomography images. The results show the effectiveness of our collaborative landmark detection framework in improving landmark detection accuracy, compared to other state-of-the-art methods.

A knowledge-based algorithm for automatic detection of cephalometric landmarks on CBCT images.

Cone-beam computed tomography (CBCT) is now an established component for 3D evaluation and treatment planning of patients with severe malocclusion and craniofacial deformities. Precision landmark plotting on 3D images for cephalometric analysis requires considerable effort and time, notwithstanding the experience of landmark plotting, which raises a need to automate the process of 3D landmark plotting. Therefore, knowledge-based algorithm for automatic detection of landmarks on 3D CBCT images has been developed and tested. A knowledge-based algorithm was developed in the MATLAB programming environment to detect 20 cephalometric landmarks. For the automatic detection, landmarks that are physically adjacent to each other were clustered into groups and were extracted through a volume of interest (VOI). Relevant contours were detected in the VOI and landmarks were detected using corresponding mathematical entities. The standard data for validation were generated using manual marking carried out by three orthodontists on a dataset of 30 CBCT images as a reference. Inter-observer ICC for manual landmark identification was found to be excellent (>0.9) amongst three observers. Euclidean distances between the coordinates of manual identification and automatic detection through the proposed algorithm of each landmark were calculated. The overall mean error for the proposed method was 2.01 mm with a standard deviation of 1.23 mm for all the 20 landmarks. The overall landmark detection accuracy was recorded at 64.67, 82.67 and 90.33 % within 2-, 3- and 4-mm error range of manual marking, respectively. The proposed knowledge-based algorithm for automatic detection of landmarks on 3D images was able to achieve relatively accurate results than the currently available algorithm.

Digital facial dysmorphology for genetic screening: Hierarchical constrained local model using ICA.

Down syndrome, the most common single cause of human birth defects, produces alterations in physical growth and mental retardation. If missed before birth, the early detection of Down syndrome is crucial for the management of patients and disease. However, the diagnostic accuracy for pediatricians prior to cytogenetic results is moderate and the access to specialists is limited in many social and low-economic areas. In this study, we propose a simple, non-invasive and automated framework for Down syndrome detection based on disease-specific facial patterns. Geometric and local texture features are extracted based on automatically detected anatomical landmarks to describe facial morphology and structure. To accurately locate the anatomical facial landmarks, a hierarchical constrained local model using independent component analysis (ICA) is proposed. We also introduce a data-driven ordering method for selecting dominant independent components in ICA. The hierarchical structure of the model increases the accuracy of landmark detection by fitting separate models to different groups. Then the most representative features are selected and we also demonstrate that they match clinical observations. Finally, a variety of classifiers are evaluated to discriminate between Down syndrome and healthy populations. The best performance achieved 0.967 accuracy and 0.956 F1 score using combined features and linear discriminant analysis. The method was also validated on a dataset with mixed genetic syndromes and high performance (0.970 accuracy and 0.930 F1 score) was also obtained. The promising results indicate that our method could assist in Down syndrome screening effectively in a simple, non-invasive way, and extensible to detection of other genetic syndromes.